Why AI co-scientists mean we'll need more researchers, not fewer
We will need more (not fewer) scientists
As AI co-scientists grow more capable, many researchers worry about being displaced. But this essay argues that science is not zero-sum: unlike most labor markets, the frontier of knowledge is unbounded. Using a geometric metaphor, the author contends that AI tools expand the volume of discovery faster than they fill it, making human scientists—who choose questions, maintain skepticism, and take ownership—more essential than ever.
The conduct of science is not zero-sum.
- Alive-in-2025
I kind of worry about new scientists minted today. When I did my phd in the 90s, the world wasn't blowing up. There wasn't a worry that some stupid llm (in the sense that it's not an intelligent being, talk among yourselves if you disagree) could make a bunch of discoveries in your area just by mass producing lots of different combinations of things and working out the implications and seeing if there were contradictions.
Will that phd matter, will your area be completely stretched in new directions in just a couple of years so you can't understand it? It's exciting and scary. Those people can't all just go to to a stupid dot com (like I did) and do something. Will there be associate profs in 3 years? Will there be tenure? We worried about this in the 90s and it was vastly less worrisome than today. I always figured I could learn a new technical field if it all went to crap. My grandfather got an agri engineering degree in the 1930s and he figured he could always find something else to do using that background.
And that was way way worse than today. Imagine if everyone you knew lost their house, their farm, their life. One grandparent was so poor they had no farm (sharecroppers), the other one lost their farm. And we were the lucky ones - they came from nothing and made it through college!
- Otterly99
Here is my two sides argument concerning AI in sciences (from a former biophysicist).
Positive side: in sciences, you need to have a lot of transverse skills such as presentation, data curation and data analysis. In biology in particular, students are never taught how to properly store data and even less how to automate the analysis. The pattern I saw was always the same: the data is stored in a convenient way for humans to read, but impossible to process via a script (think patches of columns in separate excel sheets, sometimes the labels are on the side of the value instead of on top). I also was guilty of this at the start of my PhD and had to learn it. With coding agent, you don't have that coding barrier anymore and can plot your data and even get coaching on how to properly store your data so I think this is a huge boon.
Second positive side is that you can get insight on papers which are a bit outside of your field and extract some information.
Negative side: AI also allows you to push more garbage, especially if you are working in a less experiment-heavy field, so it will get harder to sift through good and bad papers because now you can make a bad paper look good. This is honestly more of a deeper problem with sciences in general that has the unfortunate publish-or-die mindset where grants are given based on how many publications you have rather than the quality and impact of your publications. This will only amplify the problem as now you can spit out papers faster, […]